OBJECTIVE: Absence of dysplasia in the excised specimen following loop electrosurgical excision procedure (LEEP) for treatment of cervical intraepithelial neoplasia (CIN) 2/3 is an occasional finding of uncertain clinical significance. We evaluated several factors including age, liquid-based Pap (LBP) test, human papillomavirus (HPV) load before treatment, and HPV typing as predictors for absence of dysplasia. Absence of dysplasia in LEEP specimens was analyzed in terms of factors for recurrent disease after LEEP conization METHODS: In total, 192 women (mean age, 39.3+/-8.4 years; range, 24 to 70 years) with biopsy-proven CIN 2/3 were treated by LEEP conization. Age, LBP test, histological grade, HPV load, and HPV DNA typing were evaluated as possible predictors of the absence of residual dysplasia or recurrent disease. RESULTS: Of the LEEP specimens, 34 (17.7%) showed no dysplasia in preoperative biopsies from patients with proven CIN 2/3. Low HPV load (<100 relative light units [RLU]) was significantly related to the absence of dysplasia in LEEP specimens, using logistic regression. Margin involvement and high HPV load (> or =400 RLU) were significant factors for recurrence. CONCLUSION: Absence of dysplasia in LEEP specimens occurred in 17.7% of our specimens. Prediction of the absence of dysplasia in LEEP specimens was associated with low HPV load. Residual/recurrent disease after LEEP was associated with a positive resection margin and high viral load, and was not associated with absence of dysplasia in LEEP specimens. Even if there is no dysplasia in conization specimens, close follow-up for residual/recurrent disease is needed.